Skip to main content

PeakMetrics / Fact Checks

Elections & voting fact checks

Explore claims publishers have checked about elections & voting. Read their original ratings and follow the evidence.

5,412 published reviews in this topic

日本ファクトチェックセンター(JFC)Elections & voting

2万票しか取っていない蓮舫氏が当選?

Publisher’s rating: 誤り

Renho was elected in the 2025 House of Councillors election, but she received 339,311 votes—not 20,000. She ran in the proportional-representation contest, where her individual vote total ranked sixth overall.

日本ファクトチェックセンター(JFC)Elections & voting

川口市で10万票が無効に?

Publisher’s rating: 誤り

The review examined a claim that Kawaguchi City had 100,000 invalid ballots in the 2025 House of Councillors election. It says the claim spread on July 20, the election’s voting and counting day, and continued to circulate after the initial post was deleted. The review reports that the city election commission told the Japan Fact-Check Center that the claim was impossible. The commission’s figures were 5,409 invalid ballots for the Saitama constituency contest and 5,019 for the proportional-representation contest—not 100,000. The review also gives the corresponding total votes: 262,599 and 262,977. Those totals put the invalid-ballot rates at 2.06% and 1.91%, respectively. As a comparison, the review notes that Kawaguchi recorded 4,893 invalid ballots in the proportional-representation contest in the 2024 House of Representatives election, a figure it describes as comparable to the 2025 results. These reported counts directly address the claim’s location, election and alleged number.

日本ファクトチェックセンター(JFC)Elections & voting

NHK党が比例で1議席獲得すると社民党が0議席になる?

Publisher’s rating: 不正確

The review explains that House of Councillors proportional-representation seats are allocated using the D’Hondt method, so one party winning a seat does not by itself make another party lose one. The outcome would depend on the parties’ vote totals and whether they were competing for the final seat; the review says it could not be called almost inevitable in advance.

日本ファクトチェックセンター(JFC)Elections & voting

NHK党「ヤジ3回で公選法違反、私人逮捕できる」?

Publisher’s rating: 誤り

The review says election law has no rule making heckling a violation after three warnings, and that a speaker’s warning alone does not establish unlawful interference. It notes that sufficiently persistent disruption that forces an interruption may fall under the law, but the claimed three-warning rule is not provided for.

FACTLYElections & voting

YSR కాంగ్రెస్ పార్టీకి (YSRCP) ప్రస్తుతం ఉన్న ‘ఫ్యాన్’ గుర్తు స్థానంలో గొడ్డలి గుర్తును కేటాయించాలని కోరుతూ ఆ పార్టీ వ్యవస్థాపకుడు శివకుమార్ ఎన్నికల సంఘానికి లేఖ రాశారు.

Publisher’s rating: FALSE

Deutsche WelleElections & voting

In einem TikTok-Videosteht ein Reporter mit Mikrophon vor einem typisch roten Briefkasten der britischen Royal Mail. Im Hintergrund flattern Flaggen des Vereinigten Königreichs. Der Reporter fragt eine Passantin, wen sie bei den bevorstehenden Wahlen wählen würde. "Reform", antwortet die Frau. "Ich möchte mich einfach wieder britisch fühlen, nicht wahr?". Ein Nutzer kommentiert weiter unten: "Ich frage mich, wie viel sie ihr dafür bezahlt haben, dass sie das sagt."

Publisher’s rating: Diese Szene hat nie stattgefunden. Das Interview ist komplett gefälscht. Und der Reporter wurde mit Hilfe von Künstlicher Intelligenz (KI) generiert. Unten rechts im Bild gibt es einen versteckten Hinweis darauf. Hinter der Hand mit der britischen Fahne lugt ein schwaches Wasserzeichen mit dem Wort "Veo" hervor. Veo ist das neue Google-Tool für Videoerstellung. Das Acht-Sekunden-Video ist keine Ausnahme. Soziale Netzwerke werden derzeit von KI-generierten Videos geflutet, die das Aussehen und die Markenzeichen echter Nachrichtenbeiträge nachahmen. Sie suggerieren Berichterstattung von vor Ort, blenden Grafiken auf dem Bildschirm ein und verwenden starke Schlagzeilen. Doch oft sind die Inhalte frei erfunden, um Empörung zu provozieren, Meinungen zu manipulieren oder einfach, um Reichweite zu erzielen.